Simulating Rare Baseball Events Using Monte Carlo Methods in Excel and R

• Activity Number: 258
• Type: Contributed
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Sports
• Author(s): Scott Nestler*+ and Scott Billie and Michael Huber and Gabriel Costa
• Companies: Naval Postgraduate School and Military Academy at West Point and Muhlenberg College and Military Academy at West Point
• Keywords: Baseball ; Statistics ; Simulation ; Monte Carlo

Abstract

Statistics and baseball have had a long and venerable relationship throughout the history of the sport. Practically anything and everything has been recorded and tracked by avid baseball fans. The same fans debate statistics and argue which of the hallowed records will ever be broken. Websites such as Baseball-Reference.com and Baseball-statistics.com now offer data sets that include hitting, pitching, fielding, and base running events for both individual players and collective teams. Simulating certain events, such as hitting streaks or number of wins in a season, have become effective approaches to answering "Will this record ever be broken?'' One such seemingly-unbreakable record is Joe's DiMaggio's famous 56-game hitting streak of 1941. In this paper we apply Monte Carlo simulation techniques in both Microsoft Excel and the open-source statistical package R to determine the likelihood of such a rare event using actual data from the DiMaggio streak.